ABSTRACT In the course of a day we naturally make multiple shifts in our overall cognitive state and in our aims and intents. We go from sleep to awake, from internal dialogue to external communication, from relative immobility to planned complex movements. The neural activity which distinguishes these different high-level states is unknown and yet is a fundamental aspect to understanding overall cognitive processes. It is also a baseline substrate that is adversely impacted by a wide range of neuropsychiatric diseases. Our understanding of human cortical neurophysiology is almost entirely based on cognitive processes examined within the constraints of experimentally controlled tasks with time-locked, stimulus-driven behaviors. However, brain activity is fluid and continuous; much of our essential cognitive activity is not externally triggered but internally generated. The overarching goal of this research is to create platforms which allow for continuous acquisition of high-fidelity neural ensemble activity synchronized with behavioral data and contextual information to allow investigations into volitional changes in focus and intent. Data will be acquired from two groups of patients: those undergoing intracranial exploration for treatment of their epilepsy and patients implanted with multi- electrode arrays as part of the BrainGate clinical trial to restore communication and mobility to people with paralysis. Using this novel paradigm for investigation, we will explore the neural basis for changes in state which are dominated by receptive activities, internal thought and external interaction. We initially focus on motor behavior with a traditional task construction constraining the participant to watch, imagine and then attempt or actually move. We expand this to spontaneous activity in Aim 2. Finally, we move toward more general and abstracted investigations of volitional state by looking at an analogous task and spontaneous behavior with respect to language and communication (Aim 3). At each stage we will employ spectral, functional connectivity and data mining techniques to understand the neural underpinnings of these different states as well as the temporal dynamics that portend changes in state. Supporting all of these aims will be improvements and expansions in state-of-the-art human micro- and mesoscale neural recording environments to enable the study of continuous, real-time neural activity underlying state changes. Across aims we will explore the extent to which motor cortex alone contains the information which may be present in more widespread and higher order cortical regions. Our hypothesis is that motor cortex, does in fact, encode significant information about state in ways which encapsulate or summarize the information available in more wide spread regions. Most importantly, the questions inherent in this research are key to understanding human thought patterns at a fundamental level. This understanding has important implications not just for basic cognitive neuroscience but for our understanding of the processes which are altered in a wide range of neuropsychiatric disorders such as epilepsy, the dementias, depression, and psychosis. The results of these studies are also essential for the practical instantiation of effective brain-machine interfaces which would allow patients to act autonomously.